riff-api 0.1.1

Riffusion API

🎸 Riffusion API

This repo contains examples for creating music with the Riffusion API, and a tiny Python client with a typed interface.

This API is in a private beta and subject to change. Contact [e-mail] for questions.

Getting Started

Set your API key:

export RIFFUSION_API_KEY="<your-api-key>"

Install the Python client:

pip install riff-api

Make some music!

import riff_api

riff_api.generate_from_topic(
    topic="Indie pop banger about my dog Boris",
    save_to="output.wav",
)

Listen: output.wav

/topic

Run via Python client:

import riff_api

riff_api.generate_from_topic(
    topic="Indie pop banger about my dog Boris",
    save_to="output.wav",
)

Run via request (pip install requests):

import base64
import os
import requests

response = requests.post(
    "https://backend.riffusion.com/v1/topic",
    headers={
        "Content-Type": "application/json",
        "Api-Key": os.environ.get("RIFFUSION_API_KEY"),
    },
    json={
        "topic": "Indie pop banger about my dog Boris",
    },
).json()

with open("output.wav", "wb") as f:
    f.write(base64.b64decode(response["audio_b64"]))

Run via curl:

curl -X POST \
  -H "Content-Type: application/json" \
  -H "Api-Key: ${RIFFUSION_API_KEY}" \
  -d '{"topic": "Indie pop banger about my dog Boris"}' \
  https://backend.riffusion.com/v1/topic \
  | jq -r .audio_b64 | base64 -d > output.wav

The file datatypes.py shows the schema of the API.

/riff

Run via request:

import base64
import os
import requests

response = requests.post(
    'https://backend.riffusion.com/v1/riff',
    headers={
        'Content-Type': 'application/json',
        'Api-Key': os.environ.get("RIFFUSION_API_KEY"),
    },
    json={
        'prompts': [
            { "text": "chillstep pop" },
        ],
        'lyrics': (
            "Hello from outer space\n"
            "Can you hear me?\n"
            "I'm a satellite\n"
            "And I want to be by your side"
        ),
    },
).json()

with open("output.wav", "wb") as f:
    f.write(base64.b64decode(response["audio_b64"]))

Run via Python client:

import riff_api

lyrics = """
Hello from outer space
Can you hear me?
I'm a satellite
And I want to be by your side
""".strip()

response = riff_api.generate(
    prompts=[
        Prompt(text="chillstep pop"),
    ],
    lyrics=lyrics,
)

riff_api.save_audio(response.audio_b64, "output.wav")

Examples

The examples directory explains use cases of the API. To run, first pip install riff-api then execute with Python.

• 1_simple.py - Use lyrics and a sound prompt to create music.

• 2_instrumental.py - Pass no lyrics to create instrumental music.

• 3_variations.py - Hold a constant seed while modifying other parameters to create variations.

• 4_multiple_prompts.py - Use two sound prompts with different strengths to fine tune the output.

• 5_transition.py - Use two sound prompts with start/end times to create a transition.

• 6_audio_formats.py - Change the audio format of the response.

• 7_timestamped_lyrics.py - Print the timestamp of each lyric from the generation response.

• 8_topic.py - Use a single natural language sentence to generate music with the topic endpoint.

Streamlit Demo

demo_app.py is a basic web app that demonstrates using the api to create musical transitions using two sound prompts with time ranges.

Try it: https://riffusion-api-demo.streamlit.app/

Or run locally:

pip install streamlit
python -m streamlit run demo_app.py